CN106339369B - A method and system for identifying synonyms in a dataset - Google Patents

Abstract

This application discloses the synonymous word recognition methods and system of a kind of data set, this method comprises: obtaining the data set including N parts of data；All keywords in every part of data are extracted respectively；All same new words of each keyword are determined respectively；Calculate separately the same raw value between same new word corresponding to each keyword and the keyword；The high level clump of each keyword is determined respectively；To between any two keyword whether being that synonym identifies in data set, obtain corresponding thesaurus, wherein, if the similarity between the high level clump of the first keyword and the high level clump of the second keyword is not less than default similarity threshold, and the same raw value between the first keyword and the second keyword is 0, then the first keyword and the second keyword is identified as synonym.In the application, the identification process of synonym be not necessarily to be related to morphology itself relatively or dependent on specific text structure, so as to significantly promote the recognition effect of synonym.

Description

A method and system for identifying synonyms in a dataset

technical field

The invention relates to the technical field of semantic recognition, in particular to a method and system for recognizing synonyms in a data set.

Background technique

Human-computer interaction is the science that studies the interaction between systems and users. Wherein, the above-mentioned system may be various machines, and may also be computer systems and software. For example, intelligent retrieval system, semantic understanding system, etc.

Synonyms are an important part of human-computer interaction. The automatic identification of synonyms is an important part of knowledge base research. There are many ways to automatically identify synonyms. Common methods include morphological similarity-based identification, definition-based identification, and so on. The former can only identify synonyms with similar morphological forms, but cannot identify synonyms with completely different morphological forms; the latter needs to rely on a specific structural text, and if some keywords are not defined in the text, they cannot be identified. Therefore, in specific The application is very limited.

In summary, it can be seen that how to improve the recognition effect of synonyms is a problem to be solved at present.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to provide a method and system for identifying synonyms in a data set, which improves the effect of identifying synonyms. Its specific plan is as follows:

A method for identifying synonyms in a dataset, including:

Get a data set including N pieces of data, where N is a positive integer;

Extract all keywords in each data separately;

Determine all the cognate words of each keyword respectively, wherein, the cognate words of any keyword are the keywords that appear in the same data at the same time as the keyword;

Calculate the co-occurrence value between each keyword and the co-occurrence word corresponding to the keyword respectively; wherein, co-occurrence value is used to measure the probability of any two keywords appearing in the same data;

Determine the high-value word group of each keyword respectively, wherein, the high-value word group of any keyword is the word obtained by sorting all the cognate words of the keyword in descending order of cognate value. group;

Identify whether any two keywords in the data set are synonyms, and obtain a corresponding thesaurus, wherein, if the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is The degree of similarity is not less than the preset similarity threshold, and the co-occurrence value between the first keyword and the second keyword is 0, then the first keyword and the second keyword are identified as synonyms .

Preferably, the calculation formula of the co-occurrence value between any two keywords is:

E _ij =C _ij ² /(C _i ×C _j );

In the formula, C _ij represents the total number of copies of the data of the keyword K _i and the keyword K _j that appear simultaneously in the data set, and C _i represents the total number of copies of the data of the keyword K _i in the data set; C _j represents the total number of data sets in which the keyword K _j appears in the data set; E _ij represents the co-occurrence value between the keyword K _i and the keyword K _j .

Preferably, after the process of respectively determining the high-value word group of each keyword, the method further includes:

The co-occurrence index of each keyword is calculated separately, wherein the co-occurrence index of any keyword is the average value of co-occurrence values between all co-occurrence words in the high-value word group corresponding to the keyword and the keyword.

Preferably, when the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than the preset similarity threshold, the method further includes:

If the co-occurrence value between the first keyword and the second keyword is not 0, calculate the co-occurrence index of the first keyword and the co-occurrence index of the second keyword. The average value is obtained to obtain the corresponding average syngeneic index;

Calculate the ratio between the average co-generation index and the current co-generation value, wherein the current co-generation value is the co-generation value between the first keyword and the second keyword;

Determine whether the ratio is not less than a preset ratio threshold, if so, identify the first keyword and the second keyword as synonyms, if not, identify the first keyword and the second keyword as synonyms Keywords are identified as non-synonyms.

Preferably, the preset similarity threshold is 80%, and the preset ratio threshold is 10.

The invention also discloses a synonym identification system for a data set, comprising:

The data set acquisition module is used to acquire a data set including N pieces of data, where N is a positive integer;

The keyword extraction module is used to extract all the keywords in each data separately;

The cognate word determination module is used to determine all the cognate words of each keyword respectively, wherein, the cognate words of any keyword are the keywords that appear in the same data at the same time as the keyword;

The co-occurrence value calculation module is used to calculate the co-occurrence value between each keyword and the co-occurrence word corresponding to the keyword; wherein, the co-occurrence value is used to measure the occurrence of any two keywords in the same data probability;

The high-value word group determination module is used to determine the high-value word group of each keyword respectively, wherein, the high-value word group of any keyword is arranged in descending order of co-occurrence value, and the value of the keyword is the same. The word group obtained after sorting all cognate words;

The synonym identification module is used to identify whether any two keywords in the data set are synonyms, and obtain a corresponding thesaurus, wherein if the high-value word group of the first keyword and the high-value word of the second keyword are The similarity between word groups is not less than the preset similarity threshold, and the co-occurrence value between the first keyword and the second keyword is 0, then the first keyword and the second keyword are set. Two keywords are identified as synonyms.

Preferably, when the co-generation value calculation module calculates the co-generation value between any two keywords, the corresponding calculation formula is:

E _ij =C _ij ² /(C _i ×C _j );

In the formula, C _ij represents the total number of copies of the data of the keyword K _i and the keyword K _j that appear simultaneously in the data set, and C _i represents the total number of copies of the data of the keyword K _i in the data set; C _j represents the total number of data sets in which the keyword K _j appears in the data set; E _ij represents the co-occurrence value between the keyword K _i and the keyword K _j .

Preferably, the synonym recognition system further includes:

The co-occurrence index calculation module is used to calculate the co-occurrence index of each keyword after the high-value word group determination module determines the high-value word group of each keyword respectively, wherein the co-occurrence index of any keyword is The index is the average of the co-occurrence values between all co-occurrence words in the high-value word group corresponding to the keyword and the keyword.

Preferably, the synonym recognition module further includes:

an average value calculation unit, used for the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than the preset similarity threshold, and the If the co-occurrence value between the first keyword and the second keyword is not 0, calculate the average between the co-occurrence index of the first keyword and the co-occurrence index of the second keyword value to obtain the corresponding average congenital index;

A ratio calculation unit, configured to calculate the ratio between the average co-occurrence index and the current co-occurrence value, wherein the current co-occurrence value is the co-occurrence between the first keyword and the second keyword value;

A ratio judgment unit, configured to judge whether the ratio is not less than a preset ratio threshold, if so, identify the first keyword and the second keyword as synonyms, and if not, identify the first keyword The word and the second keyword are identified as non-synonymous.

Preferably, the preset similarity threshold is 80%, and the preset ratio threshold is 10.

It can be seen that, before identifying whether any two keywords in the data set are synonyms, the present invention first calculates the co-occurrence value between each keyword in the data set and the congenital word corresponding to the keyword, and determines each The high-value word group of a keyword, in which the so-called co-occurrence value is used to measure the probability of any two keywords appearing in the same data, and the high-value word group of any keyword refers to the co-occurrence value The word group obtained by sorting all the cognate words of the keyword in descending order. It can be understood that if the high-value word groups corresponding to any two keywords are relatively similar, it means that the two keywords are likely to be synonyms. The life value is 0, that is, if these two keywords do not appear in the same data, because in the same data, the creator of the data will not use two words with different forms in succession to a large extent. means the same meaning. Therefore, after finding that two keywords are likely to be synonyms through the comparison of high-value word groups, if it is further found that these two keywords do not appear in the same data, the two keywords Keywords are recognized as synonyms. Obviously, the above-mentioned synonym recognition process does not need to involve the comparison of the word form itself or rely on a specific text structure, so that the recognition effect of synonyms can be greatly improved, that is, the recognition accuracy of synonyms is improved. , and the above-mentioned method for identifying synonyms is not limited by the different types of data, and can be widely applied to various text data, which has a very broad application prospect.

Description of drawings

In order to explain the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only It is an embodiment of the present invention. For those of ordinary skill in the art, other drawings can also be obtained according to the provided drawings without creative work.

1 is a flowchart of a method for identifying synonyms in a data set disclosed in an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a synonym recognition system for a data set disclosed in an embodiment of the present invention.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

An embodiment of the present invention discloses a method for identifying synonyms in a data set. Referring to FIG. 1 , the method includes:

Step S11: Acquire a data set including N pieces of data, where N is a positive integer.

Wherein, the data set in the embodiment of the present invention includes various thematic and/or subject data obtained through network and/or manual collection, such as scientific and technological documents, patent documents, medical records, factual data, and the like.

In addition, it should be noted that the greater the number of data included in the above data set, the higher the final synonym recognition accuracy.

Step S12: Extract all the keywords in each document respectively.

Among them, the keywords in each document can be either manually indexed keywords or automatically indexed by a background system.

Step S13: Determining all cognate words of each keyword respectively, wherein the cognate words of any keyword are keywords that appear in the same document at the same time as the keyword.

In this embodiment, if the keyword A and the keyword B appear in the same document at the same time, the keyword A is called the cognate word of the keyword B. Similarly, the keyword B is also called the cognate word of the keyword A. It can be understood that the above keyword A and keyword B constitute a set of homologous word pairs.

In addition, in order to facilitate the management of cognate words, in the embodiment of the present invention, all the determined cognate words may be stored in a relational database, or stored in the form of a matrix to form a corresponding cognate word matrix.

Step S14: Calculate the co-occurrence value between each keyword and the co-occurrence word corresponding to the keyword respectively; wherein, the co-occurrence value is used to measure the probability of any two keywords appearing in the same document.

In this embodiment, the so-called co-occurrence value is used to measure the probability of any two keywords appearing in the same data, that is, to measure the probability of whether any two keywords can form a pair of co-occurring words.

If, before step S14, all the synonymous words have been saved as synonymous word matrices or stored in the above-mentioned relational database, the embodiment of the present invention may further mark each synonymous value calculated in step S14 in the above-mentioned synonymous word matrix. At the corresponding position in the new word matrix or relational database, a network of cognate words composed of cognate words and corresponding cognate values is formed.

Step S15: Determine the high-value word group of each keyword respectively, wherein, the high-value word group of any keyword is arranged in descending order according to the co-occurrence value, after sorting all co-occurrence words of the keyword. obtained word group.

In this embodiment, the high-value word group of a certain keyword refers to a word group obtained by arranging all cognate words of the keyword in descending order of cognate value.

For example, the following table 1 records the high-value word groups corresponding to the keyword "5-fluorouracil" in the A data; in addition, the following table 2 records the high-value word groups corresponding to the keyword "5-FU" in the B data.

Among them, Table 1 shows that the syngeneic value between the keywords "5-fluorouracil" and "5-FU" is 0, which means that the keyword "5-FU" does not appear in the A data. Similarly, Table 2 shows that the syngeneic value between the keywords "5-FU" and "5-fluorouracil" is 0, which means that the keyword "5-fluorouracil" does not appear in the B data.

Table I

Key words Key words cogeneration value 5-Fluorouracil Anti-tumor combined chemotherapy regimen 0.0314 5-Fluorouracil Cisplatin 0.0272 5-Fluorouracil aldehyde hydrogen folic acid 0.0197 5-Fluorouracil antineoplastic drugs 0.0015 5-Fluorouracil 5-FU 0

Table II

Key words Key words cogeneration value 5-FU Anti-tumor combined chemotherapy regimen 0.0302 5-FU Cisplatin 0.0281 5-FU aldehyde hydrogen folic acid 0.0188 5-FU antineoplastic drugs 0.0019 5-FU 5-Fluorouracil 0

Step S16: Identify whether any two keywords in the data set are synonyms, and obtain a corresponding thesaurus. If the similarity is not less than the preset similarity threshold, and the co-occurrence value between the first keyword and the second keyword is 0, the first keyword and the second keyword are identified as synonyms.

It can be understood that, the above-mentioned first keyword and the above-mentioned second key are arbitrary keywords in the data set. In this embodiment of the present invention, under the condition that the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than the preset similarity threshold, if the first keyword and the second keyword have a similarity If the co-value between them is 0, the first keyword and the second keyword are identified as synonyms.

Taking the above-mentioned Tables 1 and 2 as examples, the keyword "5-fluorouracil" in Table 1 and the keyword "5-FU" in Table 2 have very similar high-value word groups. In this case, Further comparison of the syngeneic value between the keywords "5-fluorouracil" and "5-FU", it can be seen from the above Tables 1 and 2 that the syngeneic value between the keywords "5-fluorouracil" and "5-FU" is 0. Since in the same document, the creator of the document will not use two words with different morphological forms successively to express the same meaning. Therefore, in this embodiment, the above keywords "5-fluorouracil" and "5-fluorouracil" FU" was identified as a synonym.

In addition, it should be noted that the thesaurus created in the above step S16 can be directly applied to the organization and utilization of data resources, as well as fields such as artificial intelligence.

It can be seen that, before identifying whether any two keywords in the data set are synonyms, the embodiment of the present invention first calculates the co-occurrence value between each keyword in the data set and the co-occurrence word corresponding to the keyword, and determines The high-value word group of each keyword is obtained, in which the so-called co-occurrence value is used to measure the probability of any two keywords appearing in the same data, and the high-value word group of any keyword refers to the same The word group is obtained after sorting all the cognate words of the keyword in the order of the occurrence value from large to small. It can be understood that if the high-value word groups corresponding to any two keywords are relatively similar, it means that the two keywords are likely to be synonyms. The life value is 0, that is, if these two keywords do not appear in the same data, because in the same data, the creator of the data will not use two words with different forms in succession to a large extent. means the same meaning. Therefore, after finding that two keywords are likely to be synonyms through the comparison of high-value word groups, if it is further found that these two keywords do not appear in the same data, the two keywords Keywords are recognized as synonyms. Obviously, the above-mentioned synonym recognition process does not need to involve the comparison of the word form itself or rely on a specific text structure, so that the recognition effect of synonyms can be greatly improved, that is, the recognition accuracy of synonyms is improved. , and the above-mentioned method for identifying synonyms is not limited by the different types of data, and can be widely applied to various text data, which has a very broad application prospect.

The embodiment of the present invention discloses a specific method for identifying synonyms of a data set. Compared with the previous embodiment, this embodiment further describes and optimizes the technical solution. specific:

In step S14 of the previous embodiment, it is necessary to calculate the cognate value between each keyword and the cognate word corresponding to the keyword. In this embodiment, the calculation formula of the co-generation value between any two keywords is specifically:

E _ij =C _ij ² /(C _i ×C _j );

In the formula, C _ij represents the total number of data in the data set with the keyword K _i and the keyword K _j at the same time, C _i represents the total number of data in the data set with the keyword K _i ; C _j represents the key word in the data set. The total number of documents of the word K _j ; E _ij represents the co-occurrence value between the keyword K _i and the keyword K _j .

In step S15 of the previous embodiment, the high-value word group of each keyword needs to be determined separately. In this embodiment, after the process of respectively determining the high-value word group of each keyword, it may further include: separately calculating the co-occurrence index of each keyword, wherein the co-occurrence index of any keyword is the key The average value of cognate values between all cognate words in the high-value word group corresponding to the word and the keyword.

In addition, in step S16 of the previous embodiment, the following technical solution is disclosed: the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than a preset similarity threshold, and When the co-occurrence value between the first keyword and the second keyword is 0, the first keyword and the second keyword are identified as synonyms.

Because in the same data, it cannot be completely ruled out that the authors of the data use two different keywords to express the same meaning. In order to further improve the recognition accuracy of synonyms, in this embodiment, when the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than the preset similarity threshold, It also includes the following steps S17 to S19, wherein:

Step S17: If the co-occurrence value between the first keyword and the second keyword is not 0, calculate the average value between the co-occurrence index of the first keyword and the co-occurrence index of the second keyword, and obtain the corresponding The average congenital index of ;

Step S18: Calculate the ratio between the above-mentioned average co-generation index and the current co-generation value, wherein the current co-generation value is the co-generation value between the first keyword and the second keyword;

Step S19: judging whether the above ratio is not less than a preset ratio threshold, if so, identify the first keyword and the second keyword as synonyms, if not, identify the first keyword and the second keyword as non-synonymous .

In this embodiment, the preset similarity threshold is preferably set to 80%, and the preset ratio threshold is set to 10.

Correspondingly, the embodiment of the present invention also discloses a synonym recognition system for a data set, as shown in FIG. 2 , the system includes:

The data set acquisition module 21 is used to acquire a data set including N pieces of data, where N is a positive integer;

The keyword extraction module 22 is used for extracting all keywords in each data respectively;

The cognate word determination module 23 is used to determine all cognate words of each keyword respectively, wherein, the cognate words of any keyword are the keywords that appear in the same document simultaneously with this keyword;

The co-occurrence value calculation module 24 is used to calculate the co-occurrence value between each keyword and the co-occurrence word corresponding to the keyword respectively; wherein, the co-occurrence value is used to measure the occurrence of any two keywords in the same data The probability;

The high-value word group determination module 25 is used to determine the high-value word group of each keyword respectively, wherein, the high-value word group of any keyword is arranged in descending order according to the co-occurrence value. The word group obtained after sorting all the cognate words of ;

The synonym identification module 26 is used to identify whether any two keywords in the data set are synonyms, and obtain a corresponding synonym database, wherein if the high-value word group of the first keyword and the high-value word of the second keyword If the similarity between the groups is not less than the preset similarity threshold, and the co-occurrence value between the first keyword and the second keyword is 0, the first keyword and the second keyword are identified as synonyms.

It can be seen that, before identifying whether any two keywords in the data set are synonyms, the embodiment of the present invention first calculates the co-occurrence value between each keyword in the data set and the co-occurrence word corresponding to the keyword, and determines The high-value word group of each keyword is obtained, in which the so-called co-occurrence value is used to measure the probability of any two keywords appearing in the same data, and the high-value word group of any keyword refers to the same The word group is obtained after sorting all the cognate words of the keyword in the order of the occurrence value from large to small. It can be understood that if the high-value word groups corresponding to any two keywords are relatively similar, it means that the two keywords are likely to be synonyms. The life value is 0, that is, if these two keywords do not appear in the same data, because in the same data, the creator of the data will not use two words with different forms in succession to a large extent. means the same meaning. Therefore, after finding that two keywords are likely to be synonyms through the comparison of high-value word groups, if it is further found that these two keywords do not appear in the same data, the two keywords Keywords are identified as synonyms. Obviously, the above-mentioned synonym identification process does not need to involve the comparison of the word form itself or rely on a specific text structure, so that the identification effect of synonyms can be greatly improved, that is, the identification accuracy of synonyms is improved. , and the above-mentioned method for identifying synonyms is not limited by different types of data, can be widely applied to various text data, and has a very broad application prospect.

Further, when the above-mentioned co-generation value calculation module calculates the co-generation value between any two keywords, the corresponding calculation formula is:

E _ij =C _ij ² /(C _i ×C _j );

In the formula, C _ij represents the total number of data in the data set with the keyword K _i and the keyword K _j at the same time, C _i represents the total number of data in the data set with the keyword K _i ; C _j represents the key word in the data set. The total number of documents of the word K _j ; E _ij represents the co-occurrence value between the keyword K _i and the keyword K _j .

In addition, the synonym recognition system in this embodiment may further include:

The co-occurrence index calculation module is used to calculate the co-occurrence index of each keyword after the high-value word group determination module determines the high-value word group of each keyword, wherein the co-occurrence index of any keyword is The average value of cognate values between all cognate words in the high-value word group corresponding to the keyword and the keyword.

Further, the above-mentioned synonym identification module may further include an average value calculation unit, a ratio calculation unit and a ratio judgment unit; wherein,

The average value calculation unit is used for the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than a preset similarity threshold, and the first keyword and the second keyword When the co-occurrence value between words is not 0, calculate the average value between the co-occurrence index of the first keyword and the co-occurrence index of the second keyword, and obtain the corresponding average co-occurrence index;

a ratio calculation unit, used to calculate the ratio between the average co-generation index and the current co-generation value, wherein the current co-generation value is the co-generation value between the first keyword and the second keyword;

The ratio judgment unit is used for judging whether the ratio is not less than the preset ratio threshold, if so, the first keyword and the second keyword are identified as synonyms, if not, the first keyword and the second keyword are identified as non-synonyms.

Preferably, the preset similarity threshold is 80%, and the preset ratio threshold is 10.

Finally, it should also be noted that in this document, relational terms such as first and second are used only to distinguish one entity or operation from another, and do not necessarily require or imply these entities or that there is any such actual relationship or sequence between operations. Moreover, the terms "comprising", "comprising" or any other variation thereof are intended to encompass a non-exclusive inclusion such that a process, method, article or device that includes a list of elements includes not only those elements, but also includes not explicitly listed or other elements inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The method and system for identifying synonyms in a data set provided by the present invention have been described in detail above. The principles and implementations of the present invention are described with specific examples in this paper. The method of the invention and its core idea; at the same time, for those skilled in the art, according to the idea of the present invention, there will be changes in the specific implementation and application scope. In summary, the content of this description should not be understood to limit the present invention.

Claims (8)

1. a method for identifying synonyms of data set, is characterized in that, comprises: Get a data set including N pieces of data, where N is a positive integer; Extract all keywords in each data separately; Determine all the cognate words of each keyword respectively, wherein, the cognate words of any keyword are the keywords that appear in the same data at the same time as the keyword; Calculate the co-occurrence value between each keyword and the co-occurrence word corresponding to the keyword respectively; wherein, co-occurrence value is used to measure the probability of any two keywords appearing in the same data; Determine the high-value word group of each keyword respectively, wherein, the high-value word group of any keyword is the word obtained by sorting all the cognate words of the keyword in descending order of cognate value. group; Identify whether any two keywords in the data set are synonyms, and obtain a corresponding thesaurus, wherein, if the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is The degree of similarity is not less than the preset similarity threshold, and the co-occurrence value between the first keyword and the second keyword is 0, then the first keyword and the second keyword are identified as synonyms ; And, after the process of respectively determining the high-value word group of each keyword, the method further includes: The co-occurrence index of each keyword is calculated separately, wherein the co-occurrence index of any keyword is the average value of co-occurrence values between all co-occurrence words in the high-value word group corresponding to the keyword and the keyword. 2. the synonym identification method of data set according to claim 1, is characterized in that, the calculation formula of the co-occurrence value between any two keywords is: E _ij =C _ij ² /(C _i ×C _j ); In the formula, C _ij represents the total number of copies of the data of the keyword K _i and the keyword K _j that appear simultaneously in the data set, and C _i represents the total number of copies of the data of the keyword K _i in the data set; C _j represents the total number of data sets in which the keyword K _j appears in the data set; E _ij represents the co-occurrence value between the keyword K _i and the keyword K _j . 3. The method for identifying synonyms in a data set according to claim 1, wherein the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is different. In the case of being less than the preset similarity threshold, it also includes: If the co-occurrence value between the first keyword and the second keyword is not 0, calculate the co-occurrence index of the first keyword and the co-occurrence index of the second keyword. The average value is obtained to obtain the corresponding average syngeneic index; Calculate the ratio between the average co-generation index and the current co-generation value, wherein the current co-generation value is the co-generation value between the first keyword and the second keyword; Determine whether the ratio is not less than a preset ratio threshold, if so, identify the first keyword and the second keyword as synonyms, if not, identify the first keyword and the second keyword as synonyms Keywords are identified as non-synonyms. 4. the synonym identification method of data set according to claim 3, is characterized in that, The preset similarity threshold is 80%, and the preset ratio threshold is 10. 5. A synonym recognition system of a data set, characterized in that, comprising: The data set acquisition module is used to acquire a data set including N pieces of data, where N is a positive integer; The keyword extraction module is used to extract all the keywords in each data separately; The cognate word determination module is used to determine all the cognate words of each keyword respectively, wherein, the cognate words of any keyword are the keywords that appear in the same data at the same time as the keyword; The co-occurrence value calculation module is used to calculate the co-occurrence value between each keyword and the co-occurrence word corresponding to the keyword; wherein, the co-occurrence value is used to measure the occurrence of any two keywords in the same data probability; The high-value word group determination module is used to determine the high-value word group of each keyword respectively, wherein, the high-value word group of any keyword is arranged in descending order of co-occurrence value, and the value of the keyword is the same. The word group obtained after sorting all cognate words; The synonym identification module is used to identify whether any two keywords in the data set are synonyms, and obtain a corresponding thesaurus, wherein if the high-value word group of the first keyword and the high-value word of the second keyword are The similarity between word groups is not less than the preset similarity threshold, and the co-occurrence value between the first keyword and the second keyword is 0, then the first keyword and the second keyword are set. Two keywords are identified as synonyms; And, the synonym recognition system further includes: The co-occurrence index calculation module is used to calculate the co-occurrence index of each keyword after the high-value word group determination module determines the high-value word group of each keyword respectively, wherein the co-occurrence index of any keyword is The index is the average of the co-occurrence values between all co-occurrence words in the high-value word group corresponding to the keyword and the keyword. 6. the synonym identification system of data set according to claim 5, is characterized in that, when described synonymous value calculating module calculates the synonymous value between any two keywords, corresponding calculation formula is: E _ij =C _ij ² /(C _i ×C _j ); In the formula, C _ij represents the total number of copies of the data of the keyword K _i and the keyword K _j that appear simultaneously in the data set, and C _i represents the total number of copies of the data of the keyword K _i in the data set; C _j represents the total number of data sets in which the keyword K _j appears in the data set; E _ij represents the co-occurrence value between the keyword K _i and the keyword K _j . 7. The synonym identification system of the data set according to claim 5, wherein the synonym identification module further comprises: an average value calculation unit, used for the similarity between the high-value word group of the first keyword and the high-value word group of the second keyword is not less than the preset similarity threshold, and the If the co-occurrence value between the first keyword and the second keyword is not 0, calculate the average between the co-occurrence index of the first keyword and the co-occurrence index of the second keyword value to obtain the corresponding average congenital index; A ratio calculation unit, configured to calculate the ratio between the average co-occurrence index and the current co-occurrence value, wherein the current co-occurrence value is the co-occurrence between the first keyword and the second keyword value; A ratio judgment unit, configured to judge whether the ratio is not less than a preset ratio threshold, if so, identify the first keyword and the second keyword as synonyms, and if not, identify the first keyword The word and the second keyword are identified as non-synonymous. 8. The synonym identification system of the data set according to claim 7, wherein, The preset similarity threshold is 80%, and the preset ratio threshold is 10.